CNN-Based mmWave Path Loss Modeling for Fixed Wireless Access in Suburban Scenarios

Path loss modeling of millimeter waves is regarded as one of the most challenging problems in the design of the fifth-generation (5G) mobile communication networks due to the high susceptibility to various environmental influences. We present a novel convolutional neural network (CNN)-based path loss modeling method based on a set of 28 GHz mmWave field measurements in suburban scenarios. Enhanced local area multi-scanning (E-LAMS) algorithm that provides a CNN with path loss environmental information is proposed. A new CNN structure with four subnetworks and feature-sharing layers added between convolutional layers is also proposed. The proposed methods demonstrate their superior performance over empirical models and deterministic models in terms of accuracy and complexity. The root-mean-square error of 8.59 dB has been achieved in path loss prediction in the test scenarios.

[1]  Hyeokchan Kwon,et al.  CNN training for age group prediction in an illumination condition , 2020, ICT Express.

[2]  Larry J. Greenstein,et al.  An empirical indoor path loss model for ultra-wideband channels , 2003, Journal of Communications and Networks.

[3]  Andreas F. Molisch,et al.  Millimeter-Wave Channel Measurements and Analysis for Statistical Spatial Channel Model in In-Building and Urban Environments at 28 GHz , 2017, IEEE Transactions on Wireless Communications.

[4]  Ivan Vilovic,et al.  A comparison of MLP and RBF neural networks architectures for electromagnetic field prediction in indoor environments , 2011, Proceedings of the 5th European Conference on Antennas and Propagation (EUCAP).

[5]  Theodore S. Rappaport,et al.  Investigation of Prediction Accuracy, Sensitivity, and Parameter Stability of Large-Scale Propagation Path Loss Models for 5G Wireless Communications , 2016, IEEE Transactions on Vehicular Technology.

[6]  Bo Ai,et al.  The Design and Applications of High-Performance Ray-Tracing Simulation Platform for 5G and Beyond Wireless Communications: A Tutorial , 2019, IEEE Communications Surveys & Tutorials.

[7]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[8]  Yaguang Zhang,et al.  28-GHz Channel Measurements and Modeling for Suburban Environments , 2018, 2018 IEEE International Conference on Communications (ICC).

[9]  Haralabos C. Papadopoulos,et al.  Predicting Wireless Channel Features Using Neural Networks , 2018, 2018 IEEE International Conference on Communications (ICC).

[10]  Hong Cheng,et al.  CNN-Based Indoor Path Loss Modeling with Reconstruction of Input Images , 2018, 2018 International Conference on Information and Communication Technology Convergence (ICTC).

[11]  A. Atayero,et al.  Optimal model for path loss predictions using feed-forward neural networks , 2018 .

[12]  Jing Wang,et al.  Path Loss Prediction Based on Machine Learning: Principle, Method, and Data Expansion , 2019, Applied Sciences.

[13]  Xindong Wu,et al.  Object Detection With Deep Learning: A Review , 2018, IEEE Transactions on Neural Networks and Learning Systems.

[14]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[15]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[16]  Martín Abadi,et al.  TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems , 2016, ArXiv.

[17]  Limei Xu,et al.  A Real-Time Channel Prediction Model Based on Neural Networks for Dedicated Short-Range Communications , 2019, Sensors.

[18]  Isabelle Guyon,et al.  Taking Human out of Learning Applications: A Survey on Automated Machine Learning , 2018, 1810.13306.

[19]  Hyukjoon Lee,et al.  Feature Extraction for Neural Network Wave Propagation Loss Models from Field Measurements and Digital Elevation Map , 1999 .

[20]  Theodore S. Rappaport,et al.  Overview of Millimeter Wave Communications for Fifth-Generation (5G) Wireless Networks—With a Focus on Propagation Models , 2017, IEEE Transactions on Antennas and Propagation.

[21]  Theodore S. Rappaport,et al.  Millimeter-Wave Omnidirectional Path Loss Data for Small Cell 5G Channel Modeling , 2015, IEEE Access.